Hello again,
after succesfully using dual regression (works really well) I wonder about the
exact meaning of the z-scores for the inidividual subjects obtained in step 2 of
the dual regression.
Since the methodology differs somewhat from single session ICA I wonder
wether the statement from Christian Beckmanns 2005 paper "z-scores here
reflect the amount of variability explained by the entire decomposition at each
voxel location relative to the residual noise" also holds in this context.
I assume that the z-score reflects the degreee to which this voxels time-
series correlates with the overall component time-series, adjusted for the
residual noise. Is that correct?
Thank you,
Philipp
|